titanic_knn_adaline_MLP_perceptron.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Untitled0.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "niciM1fhBZh3",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "91056be2-04f5-4380-8a19-f103f80c67b4"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "!pip install keras\n",
        "from numpy.linalg import inv\n",
        "from tensorflow.keras.layers import Dense\n",
        "from collections import defaultdict\n",
        "from keras.layers import Dense, Activation\n",
        "from keras.models import Sequential, Model\n"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: keras in /usr/local/lib/python3.7/dist-packages (2.6.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RZLtkjGLBfmc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 276
        },
        "outputId": "d67c42f4-7b54-4369-ea2d-89f39a1bc50d"
      },
      "source": [
        "train_data=pd.read_csv('train.csv')\n",
        "train_data.head()"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Braund, Mr. Owen Harris</td>\n",
              "      <td>male</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>A/5 21171</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
              "      <td>female</td>\n",
              "      <td>38.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>PC 17599</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>C85</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>Heikkinen, Miss. Laina</td>\n",
              "      <td>female</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>STON/O2. 3101282</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
              "      <td>female</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>113803</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>C123</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Allen, Mr. William Henry</td>\n",
              "      <td>male</td>\n",
              "      <td>35.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>373450</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  Pclass  ...     Fare Cabin  Embarked\n",
              "0            1         0       3  ...   7.2500   NaN         S\n",
              "1            2         1       1  ...  71.2833   C85         C\n",
              "2            3         1       3  ...   7.9250   NaN         S\n",
              "3            4         1       1  ...  53.1000  C123         S\n",
              "4            5         0       3  ...   8.0500   NaN         S\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y8i8l7-EBfx4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 258
        },
        "outputId": "720d0910-7bc6-457b-aa56-6b0ce794d015"
      },
      "source": [
        "train_data=train_data.replace(['female','male'],[0,1])\n",
        "train_data=train_data.replace(['S','C','Q'],[0,1,2])\n",
        "tarin_data=train_data.fillna(0,inplace=True)\n",
        "train_data.head()"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Braund, Mr. Owen Harris</td>\n",
              "      <td>1</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>A/5 21171</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
              "      <td>0</td>\n",
              "      <td>38.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>PC 17599</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>C85</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>Heikkinen, Miss. Laina</td>\n",
              "      <td>0</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>STON/O2. 3101282</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
              "      <td>0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>113803</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>C123</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Allen, Mr. William Henry</td>\n",
              "      <td>1</td>\n",
              "      <td>35.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>373450</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  Pclass  ...     Fare  Cabin  Embarked\n",
              "0            1         0       3  ...   7.2500      0       0.0\n",
              "1            2         1       1  ...  71.2833    C85       1.0\n",
              "2            3         1       3  ...   7.9250      0       0.0\n",
              "3            4         1       1  ...  53.1000   C123       0.0\n",
              "4            5         0       3  ...   8.0500      0       0.0\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "maNcFe9NBe6Q"
      },
      "source": [
        "Y_train=train_data[['Survived']]\n",
        "X_train=train_data[['Pclass','Sex','Age','SibSp','Parch','Fare','Embarked']]\n",
        "Y_train=np.array(Y_train)\n",
        "X_train=np.array(X_train)"
      ],
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "SynBefa_zYtA",
        "outputId": "3c6535d4-6b50-4371-f00f-07d49107b859"
      },
      "source": [
        "test_data = pd.read_csv('test.csv')\n",
        "test_data = test_data.replace(['female', 'male'], [0, 1])\n",
        "test_data = test_data.replace(['S', 'C', 'Q'], [0, 1, 2])\n",
        "test_data = test_data.fillna(0)\n",
        "X_test = test_data[['Pclass','Sex','Age','SibSp','Parch','Fare','Embarked']]\n",
        "test = pd.read_csv('gender_submission.csv')\n",
        "Y_test = test[['Survived']]\n",
        "\n",
        "X_test.head()"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Embarked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>34.5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>7.8292</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>47.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7.0000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>62.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>9.6875</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>27.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.6625</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>12.2875</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Pclass  Sex   Age  SibSp  Parch     Fare  Embarked\n",
              "0       3    1  34.5      0      0   7.8292         2\n",
              "1       3    0  47.0      1      0   7.0000         0\n",
              "2       2    1  62.0      0      0   9.6875         2\n",
              "3       3    1  27.0      0      0   8.6625         0\n",
              "4       3    0  22.0      1      1  12.2875         0"
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lKDRY6K9zYzi"
      },
      "source": [
        "model=tf.keras.models.Sequential([\n",
        "                               tf.keras.layers.Dense(7,activation='relu'),\n",
        "                               tf.keras.layers.Dense(7,activation='sigmoid'),\n",
        "                               tf.keras.layers.Dense(32,activation='relu'),\n",
        "                               tf.keras.layers.Dense(128,activation='relu'),\n",
        "                               tf.keras.layers.Dense(128,activation='sigmoid'),\n",
        "                               tf.keras.layers.Dense(2,activation='sigmoid')\n",
        "\n",
        "\n",
        "])"
      ],
      "execution_count": 121,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KKoME6FMDAAH"
      },
      "source": [
        "model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.001),\n",
        "              loss=tf.keras.losses.sparse_categorical_crossentropy,metrics=['accuracy'])"
      ],
      "execution_count": 122,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DHXr0dT2zZBs",
        "outputId": "f221179e-d826-43e8-bbc8-00e27b8dac93"
      },
      "source": [
        "print(X_train.shape,Y_train.shape)\n",
        "print(X_test.shape,Y_test.shape)\n",
        "output=model.fit(X_train,Y_train,epochs=200)"
      ],
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(891, 7) (891,)\n",
            "(418, 7) (418,)\n",
            "Epoch 1/200\n",
            "28/28 [==============================] - 1s 2ms/step - loss: 0.6721 - accuracy: 0.5982\n",
            "Epoch 2/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.6497 - accuracy: 0.6117\n",
            "Epoch 3/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.6320 - accuracy: 0.6364\n",
            "Epoch 4/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.6230 - accuracy: 0.6678\n",
            "Epoch 5/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.6084 - accuracy: 0.6801\n",
            "Epoch 6/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.6172 - accuracy: 0.6768\n",
            "Epoch 7/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.6091 - accuracy: 0.6734\n",
            "Epoch 8/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.6032 - accuracy: 0.6723\n",
            "Epoch 9/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5994 - accuracy: 0.6756\n",
            "Epoch 10/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5910 - accuracy: 0.6801\n",
            "Epoch 11/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5916 - accuracy: 0.6790\n",
            "Epoch 12/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5877 - accuracy: 0.6824\n",
            "Epoch 13/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5853 - accuracy: 0.6880\n",
            "Epoch 14/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5799 - accuracy: 0.6891\n",
            "Epoch 15/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5778 - accuracy: 0.6936\n",
            "Epoch 16/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5721 - accuracy: 0.6992\n",
            "Epoch 17/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5673 - accuracy: 0.6880\n",
            "Epoch 18/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5653 - accuracy: 0.6936\n",
            "Epoch 19/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5590 - accuracy: 0.6925\n",
            "Epoch 20/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5624 - accuracy: 0.7015\n",
            "Epoch 21/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5568 - accuracy: 0.6970\n",
            "Epoch 22/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5467 - accuracy: 0.7071\n",
            "Epoch 23/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5500 - accuracy: 0.6958\n",
            "Epoch 24/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5450 - accuracy: 0.7003\n",
            "Epoch 25/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5413 - accuracy: 0.7116\n",
            "Epoch 26/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5504 - accuracy: 0.7093\n",
            "Epoch 27/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5355 - accuracy: 0.7217\n",
            "Epoch 28/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5304 - accuracy: 0.7149\n",
            "Epoch 29/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5370 - accuracy: 0.7026\n",
            "Epoch 30/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5261 - accuracy: 0.7172\n",
            "Epoch 31/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5241 - accuracy: 0.7071\n",
            "Epoch 32/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5213 - accuracy: 0.7351\n",
            "Epoch 33/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5276 - accuracy: 0.7262\n",
            "Epoch 34/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5150 - accuracy: 0.7396\n",
            "Epoch 35/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5184 - accuracy: 0.7475\n",
            "Epoch 36/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5106 - accuracy: 0.7374\n",
            "Epoch 37/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.5036 - accuracy: 0.7565\n",
            "Epoch 38/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5021 - accuracy: 0.7565\n",
            "Epoch 39/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5030 - accuracy: 0.7587\n",
            "Epoch 40/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5028 - accuracy: 0.7430\n",
            "Epoch 41/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.5068 - accuracy: 0.7520\n",
            "Epoch 42/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4981 - accuracy: 0.7666\n",
            "Epoch 43/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4840 - accuracy: 0.7789\n",
            "Epoch 44/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4884 - accuracy: 0.7722\n",
            "Epoch 45/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4853 - accuracy: 0.7755\n",
            "Epoch 46/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4823 - accuracy: 0.7767\n",
            "Epoch 47/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4773 - accuracy: 0.7778\n",
            "Epoch 48/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4766 - accuracy: 0.7800\n",
            "Epoch 49/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4763 - accuracy: 0.7823\n",
            "Epoch 50/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4739 - accuracy: 0.7879\n",
            "Epoch 51/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4808 - accuracy: 0.7744\n",
            "Epoch 52/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4699 - accuracy: 0.7845\n",
            "Epoch 53/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4634 - accuracy: 0.7868\n",
            "Epoch 54/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4587 - accuracy: 0.7912\n",
            "Epoch 55/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4628 - accuracy: 0.7890\n",
            "Epoch 56/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4564 - accuracy: 0.7957\n",
            "Epoch 57/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4662 - accuracy: 0.7935\n",
            "Epoch 58/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4661 - accuracy: 0.8002\n",
            "Epoch 59/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4541 - accuracy: 0.7924\n",
            "Epoch 60/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4532 - accuracy: 0.8036\n",
            "Epoch 61/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4726 - accuracy: 0.7823\n",
            "Epoch 62/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4526 - accuracy: 0.7912\n",
            "Epoch 63/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4430 - accuracy: 0.8036\n",
            "Epoch 64/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4500 - accuracy: 0.7991\n",
            "Epoch 65/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4496 - accuracy: 0.7969\n",
            "Epoch 66/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4481 - accuracy: 0.8002\n",
            "Epoch 67/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4428 - accuracy: 0.7980\n",
            "Epoch 68/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4641 - accuracy: 0.7868\n",
            "Epoch 69/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4421 - accuracy: 0.7991\n",
            "Epoch 70/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4594 - accuracy: 0.7912\n",
            "Epoch 71/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4589 - accuracy: 0.7767\n",
            "Epoch 72/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4547 - accuracy: 0.8081\n",
            "Epoch 73/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4407 - accuracy: 0.8025\n",
            "Epoch 74/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4351 - accuracy: 0.8070\n",
            "Epoch 75/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4401 - accuracy: 0.8047\n",
            "Epoch 76/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4312 - accuracy: 0.8058\n",
            "Epoch 77/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4502 - accuracy: 0.8036\n",
            "Epoch 78/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4409 - accuracy: 0.8002\n",
            "Epoch 79/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4386 - accuracy: 0.8126\n",
            "Epoch 80/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4406 - accuracy: 0.7980\n",
            "Epoch 81/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4397 - accuracy: 0.7957\n",
            "Epoch 82/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4330 - accuracy: 0.8114\n",
            "Epoch 83/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4271 - accuracy: 0.8025\n",
            "Epoch 84/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4318 - accuracy: 0.8103\n",
            "Epoch 85/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4457 - accuracy: 0.7969\n",
            "Epoch 86/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4435 - accuracy: 0.8025\n",
            "Epoch 87/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4282 - accuracy: 0.8058\n",
            "Epoch 88/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4336 - accuracy: 0.8103\n",
            "Epoch 89/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4374 - accuracy: 0.8036\n",
            "Epoch 90/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4227 - accuracy: 0.8148\n",
            "Epoch 91/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4253 - accuracy: 0.8036\n",
            "Epoch 92/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4236 - accuracy: 0.8114\n",
            "Epoch 93/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4229 - accuracy: 0.8137\n",
            "Epoch 94/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4168 - accuracy: 0.8126\n",
            "Epoch 95/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4275 - accuracy: 0.8092\n",
            "Epoch 96/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4275 - accuracy: 0.8126\n",
            "Epoch 97/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4166 - accuracy: 0.8126\n",
            "Epoch 98/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4348 - accuracy: 0.8036\n",
            "Epoch 99/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4262 - accuracy: 0.8092\n",
            "Epoch 100/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4196 - accuracy: 0.8092\n",
            "Epoch 101/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4308 - accuracy: 0.8103\n",
            "Epoch 102/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4200 - accuracy: 0.8137\n",
            "Epoch 103/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4170 - accuracy: 0.8182\n",
            "Epoch 104/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4179 - accuracy: 0.8137\n",
            "Epoch 105/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4228 - accuracy: 0.8103\n",
            "Epoch 106/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4128 - accuracy: 0.8126\n",
            "Epoch 107/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4181 - accuracy: 0.8070\n",
            "Epoch 108/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4277 - accuracy: 0.8058\n",
            "Epoch 109/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4292 - accuracy: 0.8103\n",
            "Epoch 110/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4128 - accuracy: 0.8159\n",
            "Epoch 111/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4157 - accuracy: 0.8171\n",
            "Epoch 112/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4187 - accuracy: 0.8159\n",
            "Epoch 113/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4119 - accuracy: 0.8092\n",
            "Epoch 114/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4156 - accuracy: 0.8137\n",
            "Epoch 115/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4172 - accuracy: 0.8114\n",
            "Epoch 116/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4167 - accuracy: 0.8103\n",
            "Epoch 117/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4131 - accuracy: 0.8204\n",
            "Epoch 118/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4100 - accuracy: 0.8171\n",
            "Epoch 119/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4189 - accuracy: 0.8215\n",
            "Epoch 120/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4098 - accuracy: 0.8182\n",
            "Epoch 121/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4168 - accuracy: 0.8092\n",
            "Epoch 122/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4105 - accuracy: 0.8260\n",
            "Epoch 123/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4177 - accuracy: 0.8148\n",
            "Epoch 124/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4110 - accuracy: 0.8171\n",
            "Epoch 125/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4205 - accuracy: 0.8159\n",
            "Epoch 126/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4117 - accuracy: 0.8171\n",
            "Epoch 127/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4121 - accuracy: 0.8204\n",
            "Epoch 128/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4079 - accuracy: 0.8092\n",
            "Epoch 129/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4089 - accuracy: 0.8137\n",
            "Epoch 130/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4042 - accuracy: 0.8260\n",
            "Epoch 131/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4161 - accuracy: 0.8215\n",
            "Epoch 132/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4148 - accuracy: 0.8238\n",
            "Epoch 133/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4041 - accuracy: 0.8227\n",
            "Epoch 134/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4102 - accuracy: 0.8126\n",
            "Epoch 135/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4088 - accuracy: 0.8103\n",
            "Epoch 136/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4029 - accuracy: 0.8249\n",
            "Epoch 137/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4107 - accuracy: 0.8215\n",
            "Epoch 138/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4132 - accuracy: 0.8126\n",
            "Epoch 139/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4078 - accuracy: 0.8204\n",
            "Epoch 140/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4023 - accuracy: 0.8193\n",
            "Epoch 141/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4037 - accuracy: 0.8159\n",
            "Epoch 142/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4030 - accuracy: 0.8182\n",
            "Epoch 143/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4014 - accuracy: 0.8215\n",
            "Epoch 144/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4048 - accuracy: 0.8215\n",
            "Epoch 145/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4071 - accuracy: 0.8305\n",
            "Epoch 146/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4082 - accuracy: 0.8238\n",
            "Epoch 147/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4043 - accuracy: 0.8182\n",
            "Epoch 148/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4084 - accuracy: 0.8114\n",
            "Epoch 149/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4072 - accuracy: 0.8159\n",
            "Epoch 150/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4164 - accuracy: 0.8171\n",
            "Epoch 151/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4055 - accuracy: 0.8171\n",
            "Epoch 152/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3979 - accuracy: 0.8215\n",
            "Epoch 153/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4100 - accuracy: 0.8227\n",
            "Epoch 154/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4061 - accuracy: 0.8182\n",
            "Epoch 155/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4043 - accuracy: 0.8283\n",
            "Epoch 156/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3999 - accuracy: 0.8238\n",
            "Epoch 157/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4046 - accuracy: 0.8193\n",
            "Epoch 158/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4070 - accuracy: 0.8148\n",
            "Epoch 159/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4078 - accuracy: 0.8171\n",
            "Epoch 160/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4236 - accuracy: 0.8092\n",
            "Epoch 161/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4020 - accuracy: 0.8148\n",
            "Epoch 162/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3942 - accuracy: 0.8249\n",
            "Epoch 163/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3917 - accuracy: 0.8249\n",
            "Epoch 164/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3967 - accuracy: 0.8272\n",
            "Epoch 165/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3894 - accuracy: 0.8249\n",
            "Epoch 166/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3980 - accuracy: 0.8215\n",
            "Epoch 167/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3936 - accuracy: 0.8260\n",
            "Epoch 168/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3949 - accuracy: 0.8283\n",
            "Epoch 169/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4089 - accuracy: 0.8294\n",
            "Epoch 170/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4049 - accuracy: 0.8249\n",
            "Epoch 171/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3979 - accuracy: 0.8260\n",
            "Epoch 172/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3930 - accuracy: 0.8204\n",
            "Epoch 173/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3968 - accuracy: 0.8283\n",
            "Epoch 174/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3924 - accuracy: 0.8283\n",
            "Epoch 175/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3959 - accuracy: 0.8260\n",
            "Epoch 176/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3919 - accuracy: 0.8272\n",
            "Epoch 177/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.4025 - accuracy: 0.8114\n",
            "Epoch 178/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3980 - accuracy: 0.8148\n",
            "Epoch 179/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3871 - accuracy: 0.8283\n",
            "Epoch 180/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3958 - accuracy: 0.8260\n",
            "Epoch 181/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3879 - accuracy: 0.8283\n",
            "Epoch 182/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3994 - accuracy: 0.8249\n",
            "Epoch 183/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3899 - accuracy: 0.8260\n",
            "Epoch 184/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3933 - accuracy: 0.8215\n",
            "Epoch 185/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3970 - accuracy: 0.8215\n",
            "Epoch 186/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3888 - accuracy: 0.8339\n",
            "Epoch 187/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3945 - accuracy: 0.8171\n",
            "Epoch 188/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3873 - accuracy: 0.8249\n",
            "Epoch 189/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3899 - accuracy: 0.8294\n",
            "Epoch 190/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3897 - accuracy: 0.8283\n",
            "Epoch 191/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4018 - accuracy: 0.8339\n",
            "Epoch 192/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3902 - accuracy: 0.8339\n",
            "Epoch 193/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3867 - accuracy: 0.8328\n",
            "Epoch 194/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3916 - accuracy: 0.8215\n",
            "Epoch 195/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3997 - accuracy: 0.8316\n",
            "Epoch 196/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3942 - accuracy: 0.8238\n",
            "Epoch 197/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.4023 - accuracy: 0.8227\n",
            "Epoch 198/200\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.3920 - accuracy: 0.8283\n",
            "Epoch 199/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3957 - accuracy: 0.8249\n",
            "Epoch 200/200\n",
            "28/28 [==============================] - 0s 3ms/step - loss: 0.3822 - accuracy: 0.8339\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "ELmWTEJvHPIl",
        "outputId": "c0e45997-25dd-4335-e707-94af618e3d3d"
      },
      "source": [
        "plt.plot(output.history['loss'])\n",
        "plt.plot(output.history['accuracy'])\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['loss', 'accuracy'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 124,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F-lyEmlVHPra",
        "outputId": "0c8754c1-3e93-4874-925d-9a82ffdfdaf9"
      },
      "source": [
        "model.evaluate(X_test,Y_test)\n",
        "pred=[[1,0,34,1,0,54.3900,2]]\n",
        "y_pred=np.argmax(model.predict(pred))\n",
        "print(y_pred)"
      ],
      "execution_count": 125,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "14/14 [==============================] - 0s 2ms/step - loss: 0.2972 - accuracy: 0.9115\n",
            "1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G4m-F0GDTRYn"
      },
      "source": [
        "X_test=np.array(X_test)\n",
        "Y_test=np.array(Y_test)"
      ],
      "execution_count": 126,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R179DsJWEjkM"
      },
      "source": [
        "class KNN:\n",
        "  \n",
        "  def __init__(self,k):\n",
        "    self.k=k\n",
        "\n",
        "  def fit(self,X_train,Y_train):\n",
        "    self.X_train=X_train\n",
        "    self.Y_train=Y_train\n",
        "\n",
        "    self.num_of_class=len(np.unique(Y_train))\n",
        "    #print(self.num_of_class)\n",
        "\n",
        "  def nearNeighbors(self,X_test):\n",
        "    distance=np.sqrt(np.sum((X_test-self.X_train)**2,axis=1))\n",
        "    near_neighbors=np.argsort(distance)[0:self.k]\n",
        "    return near_neighbors\n",
        "\n",
        "\n",
        "  def predict(self,X_test):\n",
        "    near_neighbors=self.nearNeighbors(X_test)\n",
        "    #print(near_neighbors)\n",
        "    y_pred=np.argmax(np.bincount(self.Y_train[near_neighbors])) \n",
        "    return y_pred\n",
        "\n",
        "  def evaluate(self,X_test,Y_test):\n",
        "    num_of_correct_pred=0\n",
        "    self.X_test=X_test\n",
        "    self.Y_test=Y_test\n",
        "    y=[]\n",
        "    for i in range(X_test.shape[0]):\n",
        "      #print(X_test.shape[0],X_test[i])\n",
        "      y.append(self.predict(X_test[i,:]))\n",
        "    for i in range(len(y)):\n",
        "      if(Y_test[i]==y[i]):\n",
        "        num_of_correct_pred +=1\n",
        "        #print(num_of_correct_pred)\n",
        "    return (num_of_correct_pred /len(Y_test))*100"
      ],
      "execution_count": 127,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HmnyGqgwEpzl",
        "outputId": "91504d64-1c40-4651-fa47-1754452a3c03"
      },
      "source": [
        "print(Y_test.shape)\n",
        "print(Y_train.shape)\n",
        "Y_train=Y_train.reshape(-1)\n",
        "Y_test=Y_test.reshape(-1)\n",
        "print(Y_train.shape,Y_test.shape)\n",
        "\n",
        "knn=KNN(5)\n",
        "knn.fit(X_train,Y_train)\n",
        "accuracy=knn.evaluate(X_test,Y_test)\n",
        "print(\"accuracy :\",accuracy)"
      ],
      "execution_count": 128,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(418,)\n",
            "(891,)\n",
            "(891,) (418,)\n",
            "accuracy : 65.55023923444976\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L6EGNWAiEwzA"
      },
      "source": [
        "class Adaline:\n",
        "  def __init__(self):\n",
        "    pass\n",
        "\n",
        "  def fit(self,X_train,Y_train):\n",
        "    self.w=np.matmul(np.linalg.inv(np.matmul(X_train.T,X_train)),np.matmul(X_train.T,Y_train)) \n",
        "    \n",
        "  def predict(self,X_test):\n",
        "    y_pred=[]\n",
        "\n",
        "    Y_pred=np.matmul(X_test,self.w)  \n",
        "    for i in range(len(Y_pred)):\n",
        "      if((Y_pred[i]-Y_test[i])**2)<= 0.5:\n",
        "        y_pred.append(0)\n",
        "      elif((Y_pred[i]-Y_test[i])**2)> 0.5:\n",
        "        y_pred.append(1)\n",
        "    return y_pred    \n",
        "\n",
        "  def evaluate(self,y_pred,Y_test):\n",
        "    num_of_correct_pred=0\n",
        "    y=[]\n",
        "\n",
        "    for i in range(len(Y_test)):\n",
        "      if(Y_test[i]==y_pred[i]):\n",
        "        num_of_correct_pred+=1\n",
        "    accuracy=(num_of_correct_pred/len(Y_test))*100\n",
        "    return accuracy"
      ],
      "execution_count": 129,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fcJwSBQME4JL",
        "outputId": "f52092c4-e53e-4e15-9543-29ce9f3573fa"
      },
      "source": [
        "adaline=Adaline()\n",
        "adaline.fit(X_train,Y_train)\n",
        "y_pred=adaline.predict(X_test)\n",
        "accuracy=adaline.evaluate(y_pred,Y_test)\n",
        "print(\"accuracy =\",accuracy)"
      ],
      "execution_count": 130,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "accuracy = 62.67942583732058\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "18JN3JqrFCX4"
      },
      "source": [
        "class Perceptron:\n",
        "  def __init__(self):\n",
        "    pass\n",
        "\n",
        "  def fit(self,X_train,Y_train):\n",
        "\n",
        "    errors=[]\n",
        "    iteration=[]\n",
        "    lr=0.001\n",
        "    \n",
        "    N=X_train.shape[0]\n",
        "    self.w=np.random.rand(7,1)\n",
        "    self.b=np.random.rand(1,1)\n",
        "\n",
        "    for j in range(N):\n",
        "        \n",
        "      y_pred=np.matmul(X_train[j],self.w)+self.b\n",
        "      e=Y_train[j]-y_pred\n",
        "      \n",
        "      self.w+=(lr*(X_train[j].T*e).T)\n",
        "      self.b+=lr*e\n",
        "      \n",
        "       \n",
        "      Y_pred=np.matmul(X_train,self.w)+self.b\n",
        "     \n",
        "\n",
        "  def predict(self,X_test):\n",
        "    Y_test_pred=np.matmul(X_train,self.w)+self.b\n",
        "    Y_test_pred[Y_test_pred>0.5]=1\n",
        "    Y_test_pred[Y_test_pred<=0.5]=0\n",
        "    #print(Y_test_pred)\n",
        "    return Y_test_pred\n",
        "\n",
        " \n",
        "  def evaluate(self,X_train,X_test):\n",
        "    Y_test_pred=np.matmul(X_test,self.w)+self.b\n",
        "    Y_test_pred[Y_test_pred>0.5]=1\n",
        "    Y_test_pred[Y_test_pred<=0.5]=0\n",
        "\n",
        "    num_of_correct_pred=0\n",
        "\n",
        "    for i in range(len(Y_test)):\n",
        "      if(Y_test_pred[i]==Y_test[i]):\n",
        "        num_of_correct_pred+=1\n",
        "    accuracy=num_of_correct_pred/len(Y_test)*100\n",
        "    return accuracy"
      ],
      "execution_count": 131,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oqU1sUB4FU50",
        "outputId": "834b4bf4-4ed3-43bd-d3c0-64c46ae2da53"
      },
      "source": [
        "perceptron=Perceptron()\n",
        "perceptron.fit(X_train,Y_train)\n",
        "Y_test_pred=perceptron.predict(X_test)\n",
        "\n",
        "accuracy=perceptron.evaluate(X_train,X_test)\n",
        "print('accuracy ',accuracy)"
      ],
      "execution_count": 132,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "accuracy  61.961722488038276\n"
          ]
        }
      ]
    }
  ]
}